Prostate-specific antigen values at the time of prostate cancer diagnosis in African-American men

JAMA. 1995 Oct 25;274(16):1277-81.


Objective: To determine if African-American men with newly diagnosed prostate cancer (PC) have higher pretreatment serum prostate-specific antigen (PSA) values after adjustment for clinical stage, age, and tumor grade, and to determine if any difference detected is related to tumor volume difference.

Design: Consecutive case series of newly diagnosed PC patients between January 1990 and September 1994 and cohort analytic study of PC patients treated by radical prostatectomy (RP) and who had whole-mount pathologic tumor volume assessment between May 1993 December 1994.

Setting: Tertiary care military medical center.

Patients: A total of 541 evaluable newly diagnosed PC patients (408 white and 133 black) having pretreatment PSA assessment at one laboratory; 91 patients undergoing RP had whole-mount tumor volume analysis.

Interventions: Medical record review for pretreatment PSA value, race, tumor grade, clinical stage, and age, as well as whole-mount pathologic assessment of RP specimen and measurement of tumor volume.

Main outcome measures: The PSA differences between black and white PC patients with adjustments for age, biopsy tumor grade (Gleason score), and clinical stage (TNM stage); PSA differences between black and white PC patients undergoing RP with adjustment for age, RP grade, clinical stage, and tumor volume.

Results: The mean (geometric) PSA value for 133 black men was 14.00 ng/mL compared with 8.29 ng/mL for 408 white men (P < .001). The black patients had higher PSA values across all stage, grade, and age categories. The racial difference in PSA levels remained statistically significant when stage, grade, and age were simultaneously controlled for (P < .001). Multivariable odds ratio testing revealed that even after adjustment for stage, grade, and age, black patients were 2.2 times as likely as white patients to have a PSA value greater than 10.0 ng/mL (95% confidence interval, 1.3 to 3.6). Tumor volume (geometric mean) was 5.42 cm3 and 2.10 cm3 for black and white RP patients, respectively (P = .002). Across all clinical stages (T1a to T3), black men had tumor volumes 1.3 to 2.5 times greater than those of white men. Multivariable analysis of covariance revealed that tumor volume and stage of disease were important predictors of PSA level, but race, grade, and age were not. (The percentage of white and black patients whose cancer was detected by screening [75.4% vs 70.4%] or who had symptoms [37.7% vs 29.6%] was not significantly different.)

Conclusions: As a group, African-American men with newly diagnosed PC have higher PSA values at initial diagnosis than white men. This PSA difference appears to be due to larger tumor volumes within clinical (TNM) stage categories among black patients. Elevated PSA value was a surrogate for larger tumor volume in this cohort of black men. This stage-for-stage tumor volume disparity even in an equal-access health care environment should prompt further study of screening behavior and/or biological differences of PC in the black population.

Publication types

  • Comparative Study
  • Research Support, Non-U.S. Gov't
  • Research Support, U.S. Gov't, Non-P.H.S.

MeSH terms

  • Age Distribution
  • Aged
  • Analysis of Variance
  • Blacks*
  • Cohort Studies
  • Humans
  • Male
  • Middle Aged
  • Multivariate Analysis
  • Neoplasm Staging
  • Prostate-Specific Antigen / blood*
  • Prostatectomy
  • Prostatic Neoplasms / blood*
  • Prostatic Neoplasms / ethnology*
  • Prostatic Neoplasms / pathology
  • Prostatic Neoplasms / surgery
  • Whites


  • Prostate-Specific Antigen